Uterine prostaglandin levels following stimulation of the decidual cell reaction: Effects of indomethacin and tranylcypromine

PROSTAGLANDINS

UTERINE PROSTAGLANDINLEVELS FOLLOWINGSTIMULATIONOF THE DECIDUAL CELL REACTION: EFFECTS OF INDOMETHACIN AND TRANYLCYPROMINE' Haldor T. Jonsson, Jr., Judith C. Rankin, Barry E. Ledford and Billy Baggett Departmentof Biochemistry Medical Universityof South Carolina 171 Ashley Avenue Charleston,South Carolina 29403 ABSTRACT ProstaglandinE (PGE)andF (PGF) levels were measured in mouse uteri at various times after either trauma (hemostatcrushing)or oil stimulationof the decidual cell reaction(DCR). The oil induced DCR led to an early increase (within 5 min) in both PGE and PGF levels. Both returned to baseline by 1 h after stimulation. A second peak in PGF levels was observed at 120 min after oil stimulation. This study demonstratesa distinct differencebetween the pattern of PGE and PGF changes in the uterus followingoil stimulation of the DCR. Indomethacinpretreatmentcompletelyblocked the oil stimulatedDCR as well as all prostaglandinincreasesfollowingeither stimulus. The trauma stimulatedDCR was not completelyblocked by indomethacinpretreatment. Pretreatmentwith tranylcypromine,an inhibitorof prostacyclin biosynthesis,did not block the prostaglandinE and F increases,but did block the oil stimulatedDCR. These findingssuggest that prostacyclinmay be an early mediator of the DCR. INTRODUCTION The transformationof uterine stromal cells to decidual cells accompaniesnormal implantation(1). A decidual cell reaction (DCR) can be elicited in the hormonallyprimed ovariectomizedmouse. This is generallyaccomplishedby physical trauma or injectionof certain fluids into the lumen of the uterus (1,2). The involvementof prostaglandinsor related compoundsin implantationhas been demonstrated by the use of indomethacin,an inhibitorof prostaglandinsynthetase, in rats and mice (3,4,5,6,7). Systemic or intralwninaladministration of indomethacinhas been shown to inhibit or abolish the DCR in the rat and mouse (7,8,9,10). The indomethacinblockade of the DCR lpreliminaryresults were presented at the 60th Annual Meeting of the EndocrineSociety, 1978, Miami, Florida, and at the 12th Annual Meeting of the Society for the Study of Reproduction,1979, Quebec, Quebec, Canada.

DECEMBER 1979 VOL. 18 NO. 6

847

PROSTAGLANDINS

has been reversed to some extent by PGF2o (3,4). Furthermore, Sananes of decidualization in hormonally primed --et al. (10) reported induction prepubertal rats by intraluminal injection of PGF2,. Recently, other products of arachidonic acid metabolism have been identified in uterine tissue. Fenwick et al. (11) have shown that 6-keto-PGFlo,.the stable metabolite of prostacyclin (PG12), is the major prostaglandin produced by uterine homogenates from pseudopregnant rats. Elevated levels of 6-keto-PGFlu have been reported by Kennedy and Zamecnik (12) in uterine implantation sites of rats on day 5 of pregnancy compared to tissue from inter-implantation sites. We have previously shown that the oil stimulated DCR is associated with a lo-fold increase in cyclic AMP within 15 min after stimulation (13), whereas decidual cells cannot be detected until 30 h after stimulation (14). More recently, using the same animal model, we have shown that uterine PGF levels increase lo-fold within 5 min after oil stimulation of the DCR (7). Both the initial rise in PGF and the subsequent DCR are blocked by pretreatment with indomethacin. To investigate further the role of arachidonic acid metabolites in the DCR, PGE levels were measured at various times following stimulation. Separate experiments were undertaken to examine the effect of tranylcypromine, an inhibitor of prostacyclin biosynthesis, on the Tranylcypromine blocks the conversion of decidual cell reaction. prostaglandin endoperoxides to PG12, while other monamine oxidase inhibitors do not (15). MATERIALSAND METHODS Female ICR mice (Flow Laboratories) weighing approximately 25 g were ovariectomized. One week after ovariectomy, animals were placed on a hormonal regimen developed by Miller and Emmens (16). Animals received three daily subcutaneous injections of 0.1 pg estradiol in 0.1 ml sesame oil followed by two days without any treatment. Animals were then given daily injections of 6.7 ng estradiol plus 1 mg progesterone in 0.1 ml sesame oil for the remainder of the experiment. Six hours after the third estradiol + progesterone injection, the right uterine horn was given a decidualizing stimulus either by trauma of 10 l.11 of sesame oil (crushing with hemostats) , or by injection The effect of cholera through the ovarian end of the uterine horn. toxin was also tested by injecting 5 pg cholera toxin (in 10,nl saline) through the ovarian end of the right uterine horn. The schedule of indomethacin treatments was modified from that described Treated animals were injected with 150 ng of by Saksena et al. (4). indomethacin(cc. in 0.1 ml of sesame oil) 6 h and again 0.25 h prior Control animals received 0.1 ml sesame oil at the to stimulation. Tranylcypromine (20 mg/kg) was administered i.p. 30 min same times. Controls received saline at the same time. prior to stimulation. Animals were killed by cervical dislocation at various times and measured after stimulation of the DCR. Cyclic AMP was extracted For prostaglandin measurements, uterine as previously described (13). horns were removed and immediately frozen on dry ice. The tissues

848

DECEMBER 1979 VOL. 18 NO. 6

PROSTAGLANDINS

remained frozen on dry ice until they were processed (24-48 h). Freezing for this period of time had no measureable effect on prostaglandin levels (7). Prostaglandins E and F were extracted and separated by the method of Orczyk and Behrman (17) as modified by Jonsson and Daniel1 (18). Tracer amounts of 3H-PGE2 and 3H-PGF2, were added prior to homogenization for the determination of percentage recoveries. PGE was measured in individual uteri by radioimmunoassay using an antibody supplied by Dr. John M. Marsh (The Endocrine Laboratory, University of Miami School of Medicine, Miami, Florida). The RIA was carried out by the procedure of Clark -et al. (19). The column fraction containing PGE was dried under N2 at 35OC and redissolved in 400 ~1 of assay buffer. One-fifth of the sample (80 ul) was counted for calculation of recoveries and three-fifths (240 ul) assayed for PGE. PGF levels in individual uteri were measured using a radioimmunoassay kit obtained from Clinical Assays, Inc. (Cambridge, MA). The assay was performed according to the manufacturer’s procedure, which was a modification of previously published procedures (20-23). The column fraction containing PGF was dried under N2 at 350C and redissolved in 1 ml Tris buffer, pH 7.4. One-fifth of the sample (200 1~1) was counted for calculation of recoveries and three-fifths (600 ~1) assayed for PGF. The specificity of the PGE assay and the efficiency of the extraction procedure were established in this laboratory for uterine tissue. The recovery of PGE2 was consistently 70-80%. There was no cross reactivity with PGF2o when present in amounts 1000 times greater than those of PGE2 used in the assay. Our PGE antibody does notcrossreact with 6-keto-PGFlo, the stable metabolite of prostacyclin (Clark, personal communication). This fact may be important, in view of Fenwick’s recent report that 6-keto-PGFl elutes with the PGE fraction from the silicic acid column (11). Furt F:ermore, the two inactive metabolites of PGE (13,14-dihydro-PGE2 and its 15-keto analog) do not significantly cross react with our PGE antibody (Clark, personal communication). Data were analyzed using the Student’s t test of correlated means for small groups. RESULTS Table 1 shows that 5 min after injection of either sesame oil or mineral oil into the lumen of a hormonally primed ovariectomized mouse there was a lo-fold increase in PGE levels in the stimulated horn (p
DECEMBER 1979 VOL. 18 NO. 6

PROSTAGLANDINS

increased levels of PGE in both the oil and trauma stimulatedhorns, Intraluminalinjectionof cholera toxin, which has previouslybeen shown to stimulatethe DCR (13), did not significantlyalter PGE levels (pcO.05). It is possible to elicit a DCR with cholera toxin in animals that have been pretreatedwith indomethacin.

Table 1.

Uterine PGE levels.

Treatmenta

PGS (pg/horn) Right Left

Control

430 + 94

313 + 74

Sham Operation

300 f 50

372 f 78

2776 k 790

316 k 88

3558 f 654

385 i:65

Mineral Oil Sesame

Oil

Indomethacin+ Sesame Oil Trauma

294 f

39

218 + 45

1852 + 261

306 f 38

Indomethacin+ Trauma

210 f

50

248 f 72

Cholera Toxin

618 k 119

433 f 60

"Ovariectomizedmice were hormone primed as outlined in Materials and Methods. Indomethacin(150 ug/O.l ml sesame oil) was injected6 h and 0.25 h prior to stimulation. The right uterine horn was stimulated. Neither horn was stimulatedin control and sham operation mice. The PGE levels were measured 5 min after stimulation. Values are the means .from6 animals per group f SEM.

Figure 1 shows the levels of PGE in stimulatedand unstimulated uterine horns at various times after intraluminalinjectionof sesame oil into the right horn. The level of PGE in stimulateduteri was elevatedseveral-foldfor the first 30 min, and remained somewhat elevated for 3 h, returningto the levels of the unstimulated horns at 4 h. The level of PGE did not change significantlyin the unstimulatedhorns during the time period studied. Levels of PGF in the same uterine horns at various times after oil stimulationare shown in Figure 2. The level of PGF in the stimulateduterine horns was elevatedmore than lo-fold as comparedwith control horns at 5 and 15 min after stimulation;but had returned to the level of the unstimulatedhorn at 30 min and remained low at 60 min. There was a

850

DECEMBER 1979 VOL. 18 NO. 6

PROSTAGLANDINS

Figure 1. PGE levels followingoil stimulationof the decidual cell reaction. Ovariectomizedmice were hormone primed as outlined in Materials and Methods. Animals were killed at various times after oil stimulationof the DCR. PGE levels were measured by RIA. Values are the means from 6 animals per group; bars indicate standard errors of the means.

-alnAAlmmal --l#mlwAm

nml

I ~Py__~--+__,_____r_____r ,

L

I

3

4

mmmY*l

Figure 2. PGF levels followingoil stimulationof the decidual cell reaction. Ovariectomizedmice were hormone primed as outlined in Materials and Methods. Animals were killed at various times after oil stimulationof the DCR. PGF levels were measured by RIA. Values are the means from 6 animals per group; bars indicate standard errors of the means.

DECEMBER 1979 VOL. 18 NO. 6

851

PROSTAGLANDINS

second increase in the level of PGF in the stimulated uteri, reaching a peak about 2 h after oil stimulation. At 4 h, PGF levels in the stimulated uteri had again returned to the levels of the contralateral unstimulated horns. The second peak in PGF in the stimulated uteri is blocked by indomethacin pretreatment (Table 2). Data in Table 3 show that tranylcypromine, an inhibitor of prostacyclin biosynthesis, greatly reduced the degree of decidualization in response to the oil stimulus (pO .OS) and PGF (~‘0.05) . The increase in the CAMPlevel still occurred (p
Table

2.

The effects on uterine

of tranylcypromine and indomethacin pretreatment PGF levels 120 min after oil stimulation.

Treatmenta PGF (pg/horn) stimulated uterine horn Sesame Oil Tranylcypromine Sesame Oil Indomethacin Sesame Oil

+

+

unstimulated uterine horn

3223 It 361

375 + 72

2920 +- 432

333 f 86

278 +

53

260 + SO

aOvariectomized mice were hormonally primed as outlined in Materials Tranylcypromine (20 mg/kg body weight) was administered and Methods. Indomethacin (150 pg/O. 1 ml i.p. 30 min prior to stimulation. sesame oil, s.c.) was administered 6 h and again 0.25 prior to The right uterine horn was stimulated intraluminally stimulation. Prostaglandin F levels were measured 120 min after with sesame oil. Values are the means from 3 animals per group + SEM. stimulation.

852

DECEMBER 1979 VOL. 18 NO. 6

PROSTAGLANDINS

Data in Table 2 show that the level of PGF two hours after oil stimulation,the time of the second peak in PGF concentration,is also undiminishedby the tranylcyprominetreatment (p>O.OS). Pargyline,which shares with tranylcyprominean inhibitoryaction on monamine oxidase but is not known to influenceprostacyclinsynthesis (15), had no influenceon the DCR or on the CAMP response to intraluminaloil when given at a dosage of 50 mg/kg body weight, an effectivemonamine oxidase inhibitory-dosage(data-notshown).

Table 3. The effect of tranylcyprominepretreatmenton uterine prostaglandinand cyclic AMP levels and on uterine horn weights followingoil or trauma stimulation. Treatmenta

PGE (pg/horn) unstim stim

PGF stim(pg/horn) unstim

Sesame Oil

3328 ?:701

333 + 60

3320 + 439

312 f 43

Tranylcypromine+ Sesame Oil

2428 ? 423

367 f 32

3050 f 332

262 + 32

cyclic AMP (pmoles/horn) stim unstim

Horn Weight (mg) stim unstim

Sesame Oil

748 f

32

95%

5

279 +

14

29i

3

Tranylcypromine+ Sesame Oil

354 +

87

93*

3

58?

10

29+- 1

Trauma

311 f

45

91*

5

89f

13

2s+

2

Tranylcypromine+ Trauma

296 f

81

87 + 11

912

9

31+

3

"3variectomizedmice were hormone primed as outlined in Materials and Methods. Stim = stimulateduterine horn; unstim = unstimulated contralateralhorn. Tranylcypromine(20 mg/kg body weight) was administeredi.p. 30 min prior to stimulation. The right uterine horn was stimulated. Prostaglandinlevels were measured 5 min after stimulation. Cyclic AMP levels were measured 15 minafter stimulation. Horn weights were determined72 h after stimulation. Values are the means from 9 animals per group f SEM.

DECEMBER 1979 VOL. 18 NO. 6

853

PROSTAGLANDINS

DISCUSSION The results in this paper offer additional support for a role of a prostaglandin or a related product of cyclooxygenase action on arachidonate as a mediator of the decidualization induced by oil injection into the uterine horn of a properly hormonally prepared mouse. Very rapid and marked increases were shown in the products measured, PGE and PGF. These data, along with the previously shown inhibitory action of indomethacin on the DCR (7), point to a stimulatory effect of the oil injection on the first, and apparently ratecontrolling reaction in the transformation of arachidonate to prostaglandins and thromboxanes. It is likely that the stimulation step is that catalyzed by the cyclooxygenase, prostaglandin endoperoxide synthase, since. the initial increases in PGE and PGF are synchronous. The increase in the levels of prostaglandins are immediate, and the peaks occur within 5 minutes, preceding the peak in CAMP, which occurs at about 15 minutes (13). This temporal relationship suggests a causal role for a prostaglandin or related compound in producing a stimulatory effect on adenylate cyclase, but better support for this order of events is the demonstration that indomethacin prevents the increase in CAMPlevels resulting from oil injection (7). Evidence in this paper and in earlier ones (7,13) shows that crushing the uterine horn (trauma) or injecting cholera toxin into the lumen of the horn will induce the DCR, and that these stimuli are effective in mice previously treated with indomethacin in a regimen blocking the DCR in These stimuli do not lead to the response to luminal oil injection. early increases in PGE or PGF in the indomethacin treated mice, thus verifying the effectiveness of the blockade and showing that these prostaglandins most likely do not play an obligatory role in the early events leading to decidualization. Uterine CAMPlevels are, however, increased by these stimuli in the indomethacin treated mice. Therefore, an obligatory role of CAMPcannot be eliminated on the basis of available data. It seems likely that an increase in the CAMPlevel in the uterine horn, probably in the stromal cells of the endometrium, initiates transformation to decidual cells and the subsequent proliferation of these cells. The role of a prostaglandin or related compound is to serve as a meditor between the oil stimulus and the stimulation of increased CAMPaccumulation, a role not needed in the trauma or cholera toxin induced process. Only two of the products of the breakdown of prostaglandin endoperoxides, PGE and PGF, have been measured and both are augmented by The identhe oil stimulation of appropriately primed uterine horns. tity of the specific mediator of the CAMPresponse and of the DCR is However, we do present data suggesting that prostathus not certain. Tranylcypromine, known to block the cyclin is a likely prospect. at concentratransformation of the endoperoxide, PGH2, to prostacyclin tions not known to influence other parts of the arachidonate pathway to prostaglandins and thromboxanes, blocks the oil inddced DCR and the It, however, increase in CAMPlevels, just as does indomethacin. does not prevent the increases in PGE and PGF as does indomethacin.

854

DECEMBER

1979 VOL. 18 NO. 6

PROSTAGLANDINS

Thesefacts, in conjuction with the known influence of tranylcypromine on prostacyclin formation in other systems (15,24), point very strongly toward prostacyclin as the mediator. Our studies do not involve any approaches which would distinguish which cell types are involved in the formation of the early products induced by the oil stimulus. It does, however, seem plausible that the luminal epithelial cells first respond to the oil, and that a product produced by them may actually diffuse into the stroma, stimulating transformation and proliferation of the cells therein. Prostacyclin is an attractive candidate for such a postulated role as an intercellular messenger, in view of the convincing evidence that prostacyclin produced in blood vessel walls influences metabolic events in platelets which come in contact with these vessel walls, decreasing the aggregation of such platelets (25). Prostacyclin has been shown to be effective in elevating the concentration of CAMPin several experimental systems (25,26). Prostacyclin and its major metabolite, 6-ketoprostaglandin Fl , have been shown also to be produced in significant amounts by rat uterine tissue (11,12,27). The bimodal response in PGF levels as a function of time was surprising to us (Figure 2). It is particulary interesting that the peak at 2 h is not accompanied by a similar peak in the level of PGE. This could indicate a changing enzyme pattern in a given cell type with time following initiation of decidualization; however, a response in a different type of cell with a different intrinsic pattern of enzymes is perhaps a more likely explanation. It is possible that the stromal cells, having been acted upon by a prostaglandin-like mediator (prostacyclin perhaps) from epithelial cells, then go through a stage of increased cyclooxygenase activity as a later part of the process of transformation and proliferation. LIST OF REFERENCES

1) 21 3)

4) 5)

Finn, C. A. and D. G. Porter. The decidual cell reaction. In: The Uterus. Publishing Sciences Group, Inc., Acton, MA, 1975, p. 74. De Feo, V. J. Decidualization. In: Cellular Biology of the Uterus. (R. M. Wynn, ed.) North Holland Publishing Co., Amsterdam, 1967, p. 191. Lau, I. F., S. K. Saksena and M. C. Chang. Pregnancy blockade by indomethacin, an inhibitor of prostaglandin synthesis: its reversal by prostaglandins and pregesterone in mice. Prostaglandins 4: 795. 1973. Saksena, S. K., I. F. Lau and M. C. Chang. Relationship between estrogen, prostaglandin F2 and histamine in delayed implantation in the mouse. Acta Endocr!?‘nol. 81: 801. 1976. in the Kennedy, T. G. Evidence for a r;;ik for prostaglandins initiation of blastocyst implantation in the rat. Biol. Reprod. 1977. -16: 286.

DECEMBER

1979 VOL. 18 NO. 6

855

PROSTAGLANDINS

61 Kennedy, T. G.

7) 8)

9)

10) 11)

Prostaglandins and increased endometrial vascular permeability resulting from the application of an artificial stimulus to the uterus of the rat sensitized for the decidual cell reaction. Biol. Reprod. 20: 560. 1979. Rankin, J. C., B. E. Ledford, KT. Jonsson, Jr., and B. Baggett. Prostaglandins, indomethacin and the decidual cell reaction in the mouse uterus. Biol. Reprod. 20: 399. 1979. Castracane, V. D., S. K. Saksena,?%d A. A. Shaikh. Effect of IUDs, prostaglandins and indomethacin on the decidual cell reaction in the rat. Prostaglandins 6: 397. 1974. Tobert, J. A. A study of the possible role of prostaglandins in decidualization using a non-surgical method for the instillation of fluids into the rat uterine lumen. J. Reprod. Fert. -47: 391. 1976. Sananes, N. , E. E. Baulieu and C. Le Goascogne. Prostaglandin(s) as inductive factor of decidualization in the rat uterus. Mol. Cell. Endocrinol. 6: 153. 1976. Fenwick, L., R. L.-Jones, B. Naylor, N. L. Poyser and N. H. Wilson. Production of prostaglandins by the pseudopregnant rat uterus in vitro, and the effect of tamoxifen, with identification C$ 6;$to-pc;;taglandin Flo as a major product. Br. J. Pharmacol.

: 12)

.

Ennedv. T. G. and J. Zamecnik. The concentration of 6-ketoprostaglandin Flo is markedly elevated at the site of blastocyst imulantation in the rat. Prostaelandins 16: 599. 1978 Rankin, J. C., B. E. Ledford, an: B. Baggzt. Early involvement of cyclic nucleotides in the artificially stimulated decidual cell reaction in the mouse uterus. Biol. Reprod. 17: 549. 1977. Ledford, B. E., J. C. Rankin, R. R. Markwald and ByBaggett. Biochemical and morphological changes following artificially Biol. Reprod. stimulated decidualization in the mouse uterus. 15: 529. 1976. Gyglewski, R. J., S. Bunting, S. Moncada, R. J. Flower, and J. R. Vane. Arterial walls are protected against deposition of platelet thrombi by a substance (Prostaglandin X) which they make from prostaglandin endoperoxides Prostaglandins -12: 685. 1976. Miller, B. G. and C. W. Emmens. The effects of oestradiol and progesterone on the incorporation of tritiated uridine into the genital tract of the mouse. J. Endocrinol. 43: 427. 1969. Orczyk, G. P. and H. R. Behrman. Ovulation blockade by aspirin or indomethacin-in vitro evidence for a role of prostaglandin in gonadotropin sec%tK Prostaglandins 1: 3. 1972. Filtered levels of PGF Jonsson, H. T., Jr. and H. B. Daniell. Prostain cat spinal cord tissue following traumatic injury. glandins 11: 51. 1976. J. M. Marsh and W. P. Lemaire. Stimulation of Clark, M.R., prostaglandin accumulation in preovulatory rat follicles by Endocrinology 1978. adenosine 3’S’-monophosphate. -102: 39. I-

13) 14)

15)

.

16) 17) 18) 19)

856

DECEMBER

1979 VOL. 18 NO. 6

PROSTAGLANDINS

M., J. W. Smith, W. T. Newton and C. W. Parker, Radioimmunoassayfor prostaglandins. Science 171: 494. 1971. 211 Levine, L. and H.Van Vunakis. Antigenic activ- of prostaglandins. Biochem. Biophys. Res. Comm. 41: 1171. 1970. 22) Caldwell,B. V., S. Burstein,W. A. Brocrand L. Speroff. Radioimmunoassayof the F prostaglandins. J. Clin. Endocr. -33: 171. 1971. Hickler, R. B. ProstaglandinSymposiumof the Worchester 231 Foundationfor ExperimentalBiology (Ramwell,P. W. and Shaw, J. eds.), Interscience,N. Y. p. 279. 1968. Blajchman,M. A., A. F. Senyi, J. Hirsh, Y. Surya, M. Buchanan 241 and J. F. Mustard. Shorteningof the bleeding time in rabbits by hydrocortisonecaused by inhibitionof prostacyclingeneration by the vessel wall. J. Clin. Invest. 63: 1026. 1979. Tateson, J. E., S. Moncada and J. R. VGe. Effect of prosta25) cyclin (PGX) on cyclic AMP concentrationsin human platelets. Prostaglandins13: 389. 1977. Claesson,H. E .,J. A. Lindgren and S. Hammerstrijm.Elevation 26) of adenosine3', 5'-monophosphatelevels in 3T3 fibroblastsby arachidonicacid: Evidence for mediationby prostaglandin12. FEBS Letters 81: 415. 1977. Williams, K. ?T, A. Dembinska-Kiec,A. Zmuda and R. J. Gryglewski. 271 Prostacyclinformationby myometrialand decidual fraction of the pregnant rat uterus. Prostaglandins-15: 343. 1978.

20) Jaffee, B.

ACKNOWLEDGEMENTS This work was supportedby USPHS grant HD-06362 from the National Instituteof Child Health and Human Development. The authors express their appreciationto Ms. Mary Neal Berkaw for technical assistance. We wish to thank Dr. J. E. Pike, Upjohn Co., Kalamazoo, Michigan for providing the prostaglandinsfor the studies.

Editor: H. Behrman Received B/7/79, Accepted 11/5/79

DECEMBER

1979 VOL. 18 NO. 6

857